Re: blockchain and linked data questions

On 10 January 2016 at 11:15, Henry Story <henry.story@bblfish.net> wrote:

>
> On 10 Jan 2016, at 01:22, Melvin Carvalho <melvincarvalho@gmail.com>
> wrote:
>
>
> I did the data modeling already.  Not the DHT tho.
>
> https://w3id.org/cc
>
> My current line of thinking is around private block chains (with a slight
> twist) ... more soon!
>
>
> Nice.
>
> The ontology looks very much like a first draft though. None of the
> relations have
> domains or ranges specified in RDFS. And there is no link to documentation
> from
> the various blockchain protocols to allow one to verify the design
> decisions. For
> something like this it actually looks like OWL modelling would be quite
> important,
>  to verify that the model was consistent and did not contain
> contradictions, and to
> make sure it was used consistentlty.
>

I didnt add owl ranges because they were not needed.  The vocab is complete
and can model most block chains.  Feel free to model it yourself (I
encourage you to do so!), you'll end up in the same place.


>
> It also looks like what is missing is a peer reviewed paper that would go
> with this.
>
> Btw. I wonder if one could not use the ontologies from the web payments
> group
> https://web-payments.org/  such as digital signatures
> https://web-payments.org/vocabs/signature
>

Possibly, there's a lot of devil in the details.


>
> Anyway, something like this if peer reviewed could help bring a lot of
> clarity
> to what the block chain is, as it would make the logical side of the block
> chain
> explicit. So it looks like this is actually an (interesting) research
> topic.
>

Yes, but im not an academic, so not my focus.  Ive spoken to academics
about this, and not had any complaints.


>
> except it's less compact.
>
>
> Would it still be (much) less compact if one used a binary RDF notation?
>

Yes


> I am not sure what the latest on this is, but I found the following:
>
> http://www.w3.org/Submission/2011/03/
> http://www.rdfhdt.org/what-is-hdt/
>
>  [...]
>
>
> This of course still leaves open the question which of the new types
> of protocols should be used, given the movement in this space as
> indicated by Toni Arcieri's blog post
>  https://tonyarcieri.com/the-death-of-bitcoin
>

Bitcoin doesnt need any new protocols.  It works just fine.  It does one
job well.  Translating bitcoin to the web is an interesting idea if you
have a use case.  Making a web version of the P2P network may need
something like webDHT.


>
> As I understand the word "chain" in blockchain is quite important. Each
> element is
> linked to the previous one and the chain of signatures has to be verified.
> So if someone
> transferrred money from A to B, one would need to find the previous state
> of A's account
> by going from the head of the block chain to the previous state of his
> account. I guess this
> is the reason why folks need to have the whole blockchain available to
> them.
>

A block chain is just a linked list.  Nothing very special about it.  You
dont need the whole block chain, but it can help if you want to verify
balances independently.  Some block chains have missing blocks and continue
to work.


>
> But then there is work going on that also does not require this level of
> consistency. So
> there is research to be done in mapping out the space between the
> blockchain and simple
> document signatures, and explaining when what should be used.
>

There's lots of educational material out there.  Once you assimilate it
all, you'll see the bitcoin block chain is a very simple structure.  It's
actually not that interesting.  More interesting are the behavioral aspects
and how it is used.


>
> Btw, does anyone know if there is there a group in Europe that is already
> researching
> this space?
>
> Great brainstorming.
>
> Henry
>
>
>
>
>